Quantifying habitat provisioning at macroalgal cultivation sites

Macroalgal cultivation is expanding rapidly, and promises to contribute significantly towards future food and energy security, sustainable livelihoods, ecosystem services and habitat provisioning for a range of associated organisms globally. Habitat provisioning underpins biodiversity and ecosystem structure and functioning, supports many ecosystem services and has possible benefits to other marine industries, including enhancement of commercial fish stocks. In macroalgal cultivation, however, only recently has habitat provisioning started to be assessed at a local scale (within a farm's footprint) and with a range of different approaches. This review evaluates techniques used to quantify habitat provisioning in and around macroalgal cultivation sites, for species ranging from microorganisms to megafauna, and outlines recommendations to enable a more comprehensive ecological valuation of macroalgal cultivation in the future. The majority of information on biodiversity associated with macroalgal cultivation is associated with quantifying biofouling or pest organisms, rather than the contribution of colonising species to healthy ecosystem functioning. We suggest how better monitoring of macroalgal cultivation could enable an ecosystem approach to aquaculture (EAA) in the future. To achieve this, we highlight the need for stand-ardised and robust methods for quantifying habitat provisioning that will enable assessment and monitoring of macroalgal cultivation sites of varying scales and within different regions and environmental settings. Increased evidence for the potential habitat value of macroalgal cultivation sites will help inform and shape marine legislation, licencing and certification for macroalgal farmers and potentially reduce marine user conflicts, helping the industry to continue to grow sustainably using EAA.

could provide environmental benefits, which have been detailed alongside potential negative impacts in several recent reviews [5][6][7][8][9][10][11][12][13] (summarised in Figure 1).Many of these environmental effects relate to key ecosystem services (ESs) including climate regulation, storm protection, biogeochemical cycling and provisioning of food and habitat, or refugia to support secondary production for wild capture fisheries 6,[14][15][16][17] (Figure 1).The proposed ESs enhanced by macroalgal cultivation would support several UN sustainable development goals including: contributing to global health and well-being; providing economic growth and resilience in coastal communities; enabling responsible consumption and production; facilitating climate action and benefiting marine ecosystems. 18][7][8][9][10][11][12][13] Potential effects on ecosystem services detailed as positive (green (+)), negative (red (−)) and neutral or undetermined (blue (?)) and with habitat provisioning highlighted (dark teal).Some effects linked to habitat provisioning are indicated with an asterisk (*) to aid with clarity.Graphics are from the Integration and Application Network, University of Maryland Centre for Environmental Science (http://ian.umces.edu/image libra ry/) and BioRender (Biorender.com)supports many ESs such as food provisioning, water quality, maintenance of pest and disease control and recreation and ecotourism. 19bitat provisioning is included differently under prominent ES classification systems, and they are (1) maintaining habitats and nursery populations 20 ; (2) refugia, or nursery and migration habitat 21 ; (3) habitat heterogeneity 22 ; and (4) life cycle maintenance -nursery service. 23,24This broad range of terminology makes defining and quantifying the value of habitat provisioning difficult, with vague metrics in place to do so. 25Furthermore, as shown in Figure 1, there may be complex interactions between species, which are difficult to separate and monitor.In the broadest sense, habitat provisioning will vary based on abiotic conditions, farm location or farm type.For example, an offshore kelp farm in Europe will provide a very different potential habitat to a shallow-bottom eucheumatoid farm in Southeast Asia.All macroalgal cultivation sites, however, present their own challenges for monitoring habitat provisioning, which has until now limited their study. 17antifying the habitat provided by macroalgal cultivation has received little attention, and therefore, no economic or ecological valuations of this potential benefit have yet been made. 16,17,25In a recent review evaluating the available literature on the habitat value of bivalve and macroalgae cultivation of the 65 studies identified, only eight of these included macroalgal cultivation sites. 17This review also only included the habitat value of aquaculture sites for wild macroinvertebrate and fish populations because of a lack of information on how macroalgae cultivation affects other species 17 (e.g.microorganisms, marine mammals or seabirds etc.).Macroalgal cultivation can, however, provide habitat for a diverse array of fauna and flora, similar to that of wild macroalgal populations, 26 through the provision of novel suspended and benthic three-dimensional substrates, food and enhanced reproduction and recruitment opportunities. 5,7,13,17This in turn could potentially support secondary food production with spill-over benefits for fisheries. 16,17It is unclear, however, how effective temporary habitats of seasonal macroalgal cultivation sites would be at maintaining biodiversity after harvests and the removal of macroalgal biomass.It is also unclear whether sites will simply aggregate wild populations instead of enhancing overall population size, production and viability through reproduction and juvenile recruitment. 5,7,16,17Further issues relate to which species macroalgal farms will support, and whether these will differ from surrounding areas, thus potentially altering ecosystem dynamics or introducing invasive species. 27Better monitoring of species at macroalgal cultivation sites is therefore needed to address these concerns and determine whether this form of aquaculture supports habitat provisioning while quantifying what its ecological (and economic) value may be.
Increased recognition and valuation of the habitat provisioned by macroalgal cultivation would enable better farm design and management to optimise potential environmental benefits, mitigate potential negative impacts and contribute towards sustainable development and an ecosystem approach to aquaculture (EAA). 8,16,25,28EAA aims to design and integrate aquaculture within ecosystems to promote sustainable development, equity and ecosystem resilience while minimising any potential negative impacts. 29,30In macroalgal cultivation, EAA may guide policy, financing and certification schemes towards promoting increased sustainable practices in mariculture development. 11Accordingly, there is the potential for macroalgal aquaculture to lead the way as an example of sustainable EAA; however, more quantitative evidence on a wider range of the potential environmental benefits is needed, including habitat provisioning. 25is review aims to: (1) summarise evidence relating to how macroalgal cultivation could provision habitat for species spanning from microorganisms to megafauna, and (2) generate recommendations for standardised monitoring of habitat provisioning in and around macroalgal cultivation sites for these species.In turn, this could support the development and optimisation of practices for EAA, which will enable the ecological and economic value of macroalgal aquaculture to be assessed more holistically and help to inform its legislation and regulation in future.Where information on macroalgal cultivation is lacking, we draw upon some relevant studies from wild macroalgae populations and shellfish/finfish aquaculture to help guide in developing universal standardised monitoring techniques.
Adopting this approach, we also hope to seek ways of standardising the monitoring of habitat provisioning between aquaculture species, which will be particularly useful given the increasing implementation of integrated multitrophic aquaculture (IMTA) systems.

| P OTENTIAL HAB ITAT PROVIS IONING OF MACROALG AL CULTIVATI ON FOR D IFFERENT TA XONOMI C AND FUN C TIONAL G ROUPS AND ME THODS ENAB LING ITS QUANTIFIC ATION
As Figure 1 illustrates, complex interactions exist in macroalgal cultivation sites between species and between species and the environment, which will differ depending on farm scale, location and type (discussed further in Section 3).These complexities are more explicitly detailed in other recent reviews (e.g.Refs.10,17), and in this section, we outline some of the ways macroalgal cultivation may affect different taxonomic and functional groups based on previous studies, why they are important to quantify and challenges relating to monitoring.We have grouped organisms based on their taxonomic and functional groups, how they inhabit or interact with macroalgal cultivation sites, and their monitoring requirements.

| Microorganisms (Bacteria, viruses, archaea, fungi, oomycetes and protists)
][33][34][35] Macroalgae host diverse associations of microorganisms that facilitate the health and function of the host plants, such as morphological development, disease protection and antifouling properties from epibionts [36][37][38] (see Section 4).As such, macroalgae and their microbiomes should be considered as synergistic ecological units or holobionts. 36,38Additionally, the microorganisms associated with macroalgae are often sources of novel compounds that have industrial applications, for example, bacteria hosted on Eucheuma species produce enzymes that could be used in biofuel production. 392][43][44][45][46][47] Oomycete pathogens can also be responsible for extensive damage, as seen in Korean Pyropia seaweed farms, 48 and are more diverse and geographically widespread than currently acknowledged, posing threats to macroalgal cultivation in Asia and Europe. 49,50The presence of harmful microorganisms and diseases at macroalgal cultivation sites also pose the risks of spreading to neighbouring wild macroalgal populations, with the potential to cause substantial ecological damage. 51A greater understanding of the microbiome of macroalgal cultivation is therefore needed to determine both its ecological value in macroalgal cultivation, 8 and understand how the macroalgal microbiome is regulated by other associated biodiversity present at cultivation sites, which may mitigate disease outbreaks. 35,52Macroalgal cultivation sites may also potentially affect microorganisms in the water column and benthic sediments, through production of soluble dissolved and particulate organic matter, 53 deposition of detritus and the potential attraction of waste-producing species such as fish (Figure 1).
Microorganisms in the surrounding environment, however, are unlikely to be affected to the extent seen in other aquaculture species, particularly fishfarms, 54 as waste production is comparatively low (although less is known about the amounts produced at larger offshore sites).Determining interactions between macroalgal cultivation sites and microbial communities in the water column and benthos will be an important step in future research, given that microorganisms can act as powerful indicators monitoring ecosystem health. 54Relatively little research has been conducted so far to investigate this.
Microorganisms can be quantified either from the surface of the macroalgae or from water or sediment samples in macroalgal cultivation sites, which can host distinct microbiomes from one another. 34The diversity and community composition of microorganisms, such as fungi, can vary between tissue types (e.g.stipe, holdfasts and blades) of the same macroalgae, 55 so multiple samples from individuals should be collected to accurately capture their microbial diversity.Various methods can be used for microorganism quantification (Table 1), including microscopy, cell counts and RNA or DNA sequencing methods. 35,38,54,56The main constraints for quantifying macroalgal farm microbiomes, however, are generally a lack of knowledge of the microbial ecology of these systems and because currently many are not culturable by common microbial methods 34,35,41,45,46 ; however, the advancement of molecular methods may help to mitigate these issues 38,55 (Table 1).Additionally, eukaryotic microorganisms, including protists, are especially understudied within the seaweed holobiont, so particular focus needs to be more directed on determining their importance.If sampling effort of microorganisms in wild and cultivated macroalgal populations was increased, reference libraries should be compiled to make detecting and quantifying their abundance easier in future.
It should also be recognised that better understanding of the dynamics and plasticity of these microbial communities is needed, as there will be similarities in their ecological and functional roles within ecosystems.

| Plankton (pico to macro)
Plankton provide important primary food sources in marine food webs, and regulate nutrient, carbon and oxygen cycles in the oceans.Therefore, plankton abundance and diversity are important measures of ecosystem productivity and health.Macroalgal cultivation sites may provide zooplankton with shelter and food 57 ; however, plankton may also negatively affect cultivated macroalgae.For example, 'diatom felt' caused by settling diatoms has been shown to result in algal bleaching and significant economic losses for farmers in Pyropia farms in Korea. 48Macroalgal cultivation sites may also benefit overall ecosystem health by mitigating eutrophication and harmful algal blooms (HABs) through improving water quality, stabilising the water column and assimilating excess nutrients 48,[58][59][60] (Figure 1).The installation of macroalgal farms in China has been observed to alleviate eutrophication and ocean acidification, reduce turbidity and subsequently enhance phytoplankton diversity and biomass. 61,620][71] Small-to mediumscale (0-50 lines × 200 m) 7 macroalgal farms, however, are unlikely to have major impacts on phytoplankton assemblages as phytoplankton in most cases will likely pass through the site with current flow. 7,16Farms should be sited in suitable locations with sufficient nutrient concentrations and tidal mixing, and therefore with adequate environmental carrying capacity (the maximum biomass of a farmed species that can be supported without exceeding the maximum acceptable impacts to the farmed stock and its environment) 72,73 (see Section 3.5).Indeed, no significant changes in plankton were predicted or detected in models of hydrodynamic and biogeochemical processes in United Kingdom and Dutch small-or large-scale macroalgal farms. 71,74Thus, the effects of macroalgal cultivation on plankton assemblages are site-specific Water sampling, plankton nets or trawls Microorganisms and plankton Inexpensive, simple and highly replicable (+) Species can then be identified by several methods e.g.: • Microscopy -time-consuming and requires taxonomic expertise, but captures total biodiversity at a given time point (−/+) • Fluorescence microscopy -a fast and direct detection protocol for microorganisms (+) • Fluorometry -rapidly measures chlorophyll a concentrations as proxies for productivity and phytoplankton biomass (+).Does not identify species or detect zooplankton species (−) Quantifying plankton assemblages associated with macroalgal cultivation is still in its infancy. 7Standardised in situ monitoring methods commonly deployed in shellfish and finfish farms could, however, be applied and adapted to better monitor plankton within macroalgal farms 75 (Table 1).This will help to generate more accurate models to quantify habitat value of macroalgal cultivation for plankton assemblages and how it may contribute to other ESs related to HAB mitigation and bioremediation.Better determination of plankton assemblages in macroalgal cultivation sites will also be useful to detect early larval stages of species that may settle on the macroalgae, such as some epibionts 76 (see Section 2.3).

| Epibionts
in turn benefitting shellfish growth, 92 enhancing primary production, 93,94 providing protection from predation, 95,96 encouraging settlement of commercially farmed shellfish 97,98 and mitigating for disease risk. 99Epibionts also provide food sources for higher trophic level species, increasing the habitat value of a cultivation site and secondary production 78 ; however, grazing interactions on epibionts are not currently well understood.
Various studies have found similar or higher levels of epibiont biodiversity associated with cultivated kelps compared to wild populations, which suggests that suspended macroalgal farms could provide novel habitat for epibionts. 100,1013][104] Therefore, established, standardised methods are needed to quantify epibionts effectively at different sites.More targeted assessments are also needed to quantify environmentally beneficial epibionts as well as detrimental species to crop production, which have been the focus of studies to date.A better understanding of how epibionts may affect or interact with the environment will also enable a more ecosystem directed view for future development of EAA.
Census techniques for quantifying epibionts are relatively straightforward compared to other farm-associated biodiversity (Table 1), as most species are slow moving or sessile so they can be identified and enumerated directly from macroalgal biomass samples. 76,77,86,105,106Previous studies assessing epibiont diversity on macroalgal farms have generally focussed on either the holdfast (e.g. Ref. 105) or the blade (e.g.Refs 100-104) separately however, rather than as one sampling unit.Quantifying the total epibiont assemblage is required if the potential habitat value of macroalgal cultivation is to be fully evaluated.

| Benthos
Benthic or seabed communities are comprised of many bioindicator species that signal environmental health and ecosystem functioning, and thus, benthic habitat monitoring is a crucial part of environmental impact assessments supporting aquaculture licence applications. 107Nevertheless, limited research has been conducted to assess the ecological status below macroalgal cultivation sites compared to the aquaculture of finfish and shellfish.be at locations where the water depth is at least twice the depth of the cultivation infrastructure and placed in areas where the minimum water flow rates are >0.05m s −1 . 109,110Compared to longline systems that can be deployed offshore, 111 there is often little if any flexibility for cultivation depth for tropical off-bottom farming sites. 9Clearly, a better understanding of benthic interactions will help inform appropriate siting of farms in suitable environmental conditions to help mitigate for any potential negative effects on the benthos.
Ecosystem models of intensive kelp cultivation scenarios have indicated minimal effects on benthic food webs, or the potential to alter them through the provision of habitat, food and detritus. 112fects on benthic species vary between macroalgal cultivation sites globally.In tropical waters, where shallow, bottom-growing macroalgal cultivation techniques are favoured, significant changes to neighbouring benthic habitats such as corals and seagrass beds and their associated assemblages are often reported (as reviewed in Ref. 10).In temperate waters, where suspended macroalgal cultivation is favoured, benthic impacts tend to be less severe or even negligible (e.g.Refs.100,101); however, this will depend on water flow rates through the cultivation area and light penetration in the water column.In Sandu Bay, China, however, sedimentary acid volatile sulphide content (linked to lower benthic biodiversity) was greater below kelp cultivation lines than at control sites. 113This emphasises that effects on the benthos need to be evaluated on a site-specific basis, as they will be highly dependent on local environmental conditions and farm type.
Key indicators of benthic habitat health include sediment biogeochemistry (e.g.particle size, nutrient, heavy metal, oxygenreduction-potential, carbon and organic matter content) and the biodiversity, composition and abundance of benthic infauna and epifauna.Infauna relates to organisms living in the sediment, whereas epifauna relates to organisms living on the seabed.Here, we consider benthic infauna and epifauna as two separate faunal categories, due to the different census techniques required to study them.We focus on both meiofauna (45 µm to 1 mm) and macrofauna (>1 mm) due to their important ecological roles and similar quantification methods. 114,115

| Benthic infauna
Benthic infauna are comprised primarily of detritivores, grazers and filter feeders, such as polychaetes, flatworms, gastropods and bivalves, that all play key roles in recycling nutrients, filtering water and providing prey to epibenthic species. 114,116The diversity and abundance of infaunal assemblages are used as bioindicators of contamination, eutrophication and hypoxia, due to the varying tolerances of species in the community. 116,117It is therefore important to assess how macroalgal farms influence infaunal assemblage structure and function, to monitor the health of the cultivation site and of the wider habitat.Previous studies in offbottom seaweed farms in Tanzania have found reduced infaunal biomass 118 or that the infaunal assemblages more closely resemble unvegetated areas rather than seagrass beds. 119In contrast, at a Swedish longline farm, increases in infaunal species diversity and abundance have been found, indicating a positive effect of the farm on benthic health. 101The marked differences in farming systems and environmental conditions between these locations (discussed further in Section 3.1) highlights how the effects of macroalgal cultivation on benthic infauna need to be evaluated in more detail.Sampling benthic infauna generally involves taking a sediment grab or core of the seabed and determining its associated fauna and biogeochemical properties 100,101,120,121 (Table 1).
Benthic survey designs and approaches to the subsequent analysis of infaunal communities can vary greatly, however (Table 1).
Standardised monitoring protocols, methods and analyses to quantify impacts from aquaculture on benthic infauna may help to regulate benthic monitoring between aquaculture types and locations in future.

| Benthic epifauna
Benthic epifauna includes macroinvertebrates (>1 mm), such as echinoderms (e.g.sea stars, urchins and sea cucumbers), crustaceans (e.g.lobsters and crabs) and benthic fish species (e.g.flatfish), many of which are of commercial or ecological importance.Wild kelp populations contribute on average US$48,600 to $141,000 ha −1 yr −1 to capture fisheries across the four major kelp genera globally, with nine of the top 10 valuable species being benthic invertebrates such as lobsters, abalone, urchins and gastropods. 1224][125] At a small temperate kelp farm in Sweden however, no effect on benthic macrofauna was reported, 101 whereas in Tanzania, lower abundances of macrofaunal or increases in sea urchin species, that could threaten to graze on the cultivated seaweed, were found. 118Differing effects on epifauna reflect the diverse nature of macroalgal cultivation sites worldwide and highlights the need to increase monitoring across different systems.
Epifaunal assemblages may be more challenging to accurately quantify than infauna, as epifauna tend to be more mobile, patchily distributed and can also be cryptic. 126Monitoring mobile epibenthic macrofauna (>1 mm) may however use similar methods to those used to quantify pelagic fish species (Section 2.5) or marine mammals, seabirds and reptiles (Section 2.6) and these have been summarised and reviewed in Table 1.

| Finfish
Finfish species include commercially important species for fisheries or for supporting wider food webs.Macroalgal cultivation sites could provide novel habitat for fish species by offering spawning substrate, shelter and food in the form of farm biomass or epibionts, similar to wild macroalgal populations (reviewed in Refs 10,17,127,128)   (Figure 1).Potential benefits to fish species will however depend on farms being sited appropriately, to not replace natural nursery habitats such as seagrass beds, and the habitat complexity created by cultivation sites compared to what was in the area previously. 10,127,129Mariculture infrastructure may also restrict fishing activities in an area, indirectly benefitting fish populations, 130 with potential spill-over benefits for fisheries. 16Conversely, however, if macroalgal cultivation sites are poorly managed and regulated, they may act as 'ecological traps' whereby fish are attracted to farm infrastructure and become more vulnerable to capture from unregulated fisheries or natural predators, such as seals. 10,11,17,131It is also unclear whether macroalgal farms enhance juvenile recruitment or simply aggregate existing adult fish populations. 16The increased availability of macroalgae and novel epibiont prey on cultivated macroalgae may also alter fish assemblage composition.Illustrating this, increased herbivorous fish biomass was reported in farms in Tanzania compared to neighbouring seagrass beds. 129Increases in herbivorous fish may be detrimental to farm yields, or conversely beneficial to fisheries in the area 129,132 ; however, it is unclear as to how changes in fish assemblage composition will affect wider ecosystem function.
Indeed, fish assemblages in Eucheuma farms in the Philippines were found to be significantly different to those in neighbouring marine protected areas and coral reefs, with more invertivores and fewer large herbivorous fish found in farms and greater biomass and diversity found in neighbouring coral reefs. 133Conversely, multispecies macroalgal farms in Costa Rica attracted a larger number of fish species compared to control sites. 78This highlights that the effects of macroalgal cultivation on fish populations can be highly variable between sites 132 and demonstrates the importance of establishing comparative reference sites in survey design (discussed further in Section 3.2).
Finfish species are generally highly mobile and can vary significantly in relation to timing of fish reproductive cycles, which makes accurately surveying their populations challenging.For example, at an Irish seaweed farm, juvenile mackerel and pollack that were found to be abundant in summer months were absent by September, whereas wrasse, which are more associated with the benthos, remained abundant below the site across the whole study period. 134shore fish assemblages are also highly influenced by tides, with greater abundances observed during high tides. 135,136Timing of seaweed harvesting is also important for monitoring finfish populations.For example, in wild kelp beds in Norway, juvenile gadoids were 92% less abundant in harvested areas compared to unharvested areas and remained 85% lower in areas one year after harvest. 137Therefore, standardised medium-to long-term monitoring at multiple spatial scales is required to reliably assess the habitat value of macroalgal cultivation on fish populations and to inform better operational management practices and farm design.Various methods (Table 1) have been used to census fish in macroalgal cultivation sites, many of which are similar to those discussed in epibenthic surveys (Section 5.2).Models of macroalgal cultivation impacts on fish species may aid understanding of ecosystem-wide implications, but should be combined with long-term in situ studies to ground-truth model outputs. 128

| Marine mammals, reptiles and seabirds
Marine mammals, reptiles and seabirds include many species of high conservation importance, so understanding how to manage marine industries to maximise environmental protection and minimise disturbance for these groups is hugely important. 138These species also play major ecological roles 138 and are of particular value to marine industries such as tourism due to their charismatic value. 139Here we have grouped these species as they will likely interact with macroalgal cultivation sites in similar ways as they share many life-history traits, and will also require similar surveying techniques to monitor (Table 1).
Potential increases in prey species of fish and macroinvertebrates in macroalgal farms could provide foraging grounds for mammals such as seals and dolphins, reptiles such as turtles, and seabirds, which are frequently observed around finfish and shellfish aquaculture sites (e.g.Refs.140-146).Farm infrastructure, however, may interfere with the ability for mammals such as dolphins to aggregate fish prey and therefore affect their behaviour and habitat use (as discussed in Ref. 128).Macroalgal cultivation sites may also displace other mammals, reptiles and seabirds due to farm construction and operation activities (e.g. as seen in shellfish aquaculture [147][148][149] ), which could lead to malnutrition if animals are displaced from their foraging grounds. 10Minimal risks to cultivated macroalgae are perceived from vertebrate predators (unlike other aquaculture species), and they may instead help to maintain trophic balance and control grazing species as seen in wild populations. 150Herbivorous species such as green turtles (Chelonia mydas, Linnaeus, 1758), however, have been found to consume some cultivated macroalgae, 151 which may attract them to cultivation sites, causing conflicts between turtles and farmers or increasing entanglement risk. 10,152,153Other species are also susceptible to entanglement risk with farming infrastructure, as seen with stationary fishing gear 154,155 ; or some mussel lines. 156,157Lethal entanglement of critically endangered dugongs (Dugong dugon, Lacépède, 1799) has been reported in cultivation sites in the Philippines. 158Entanglement risk is generally well understood however, and can be mitigated by increasing line thickness and tautness and avoiding placement in areas of known importance, such as migratory corridors. 7,156Nevertheless, there are still considerable uncertainties about how large-scale farms that occupy large surface areas of coastal seas or entire bays influence marine mammals, reptiles and seabirds.
Census of marine mammals, reptiles and seabirds is challenging due to their generally low population densities and highly mobile existence.Furthermore, surveys need to be conducted over long timescales to fully assess impacts of cultivation sites on their populations, because many have long lifespans and slow population growth. 159Monitoring of these species therefore requires immense survey effort or specialist behavioural knowledge, and is usually conducted from land, boats or air, often relying on species breaching or being at the surface of the water to be visible (Table 1).Currently, there are very limited studies on the interactions of marine mammals, reptiles or seabirds with macroalgal farms 5,10 and further research to address this knowledge gap is essential to enable better management and enhance potential habitat benefits of cultivation sites.Recent reviews have been published regarding census around marine renewable energy infrastructure (e.g.Ref. 160) and other aquaculture types, 128 which can inform monitoring in macroalgal cultivation sites (Table 1).

| G ENER AL RECOMMENDATI ONS ON S TUDY DE S I G N FOR MONITORING HAB ITAT PROVIS IONING IN MACROALG AL CULTIVATI ON S ITE S
This review illustrates that it is not only the monitoring approach used that is important for quantifying habitat provisioning in macroalgal cultivation sites but also how and when these surveys are carried out.Standardised approaches in study design and implementation will ensure monitoring is more directly comparable across farm sites and scales, including IMTA and offshore systems.The effectiveness of different survey methods will differ between farm types, however, which we discuss further below (Section 3.1).
Here, we provide the first steps in standardising monitoring protocols for assessing habitat provisioning by macroalgal cultivation, with regard to accounting for differences in farm types, establishing reference sites, timing surveys appropriately and defining the necessary species metrics to be taken for relevant habitat value analysis.
We also discuss how the collection of appropriate field data can be used in models to answer some of the broader, ecosystem-wide effects of the habitat value of macroalgal cultivation that extend beyond the footprint of the farms, and importantly how these data can be shared in open-access repositories to advance the sustainable design of macroalgal cultivation sites in EAA.

| Accounting for differences in macroalgal cultivation farm types
Globally, there is a wide diversity of macroalgal farm types and species cultivated, from offshore temperate kelp farms to off-bottom tropical carrageenophyte cultivation (Figure 1).This diversity makes creating standardised monitoring techniques to quantify habitat provisioning challenging, as their effectiveness will differ widely between cultivated species, locations and farm types.From the monitoring techniques outlined in Table 1, many of these are applicable to all macroalgal cultivation sites globally.For example, diver-conducted visual surveys of fish and pelagic species have been used successfully in both Eucheuma and Kappaphycus farms in the Philippines 133 and Codium, Gracilaria, Sargassum and Ulva farms in Costa Rica. 78This survey method is straightforward and flexible due to divers or snorkelers being able to adapt their positions around farming infrastructure.Between farm types, where infrastructure differs greatly however, some monitoring techniques are not feasible to deploy universally.For instance, small beam trawls may be suitable for monitoring benthic species around kelp longline systems (as conducted in a Canadian mussel farm 187 ); however, they are not suitable in most tropical macroalgal farms due to shallow depth limitations, high density of cultivation lines or use of mesh nets to seed (e.g. in Pyropia or Ulva farms). 9,10To survey benthic species in shallow tropical farms, more targeted methods could be deployed, such as benthic drop cameras or traps (Table 1), which can also be used successfully in temperate systems. 101Remote camera surveys would also cause less disturbance to both biodiversity and the cultivated species.To enable more effective comparison of habitat value between macroalgal cultivation sites globally, monitoring methods that are flexible in terms of their deployment should be favoured over those that will only work in certain cultivation scenarios.
Working collaboratively with farmers or other stakeholders, such as local fisheries, will enable suitable survey methods for the region to be developed.For example, in Tanzanian off-bottom Eucheuma and Kappaphycus farms, where trawling or netting is not possible, researchers used traditional 'madema' basket traps to survey fish populations, and were instructed on best deployment techniques by local fishermen. 129Using local methods and knowledge-enabled successful catches of fish populations in relation to macroalgal cultivation sites and also to assess how the presence of macroalgal cultivation sites may affect the economic value of other marine industries in the area. 129Engaging with farmers and other local industries may also help to improve awareness of the potential ecological and economic benefits of habitat provisioning by macroalgal cultivation 9 and therefore increase interest in contribution to farmer or citizen science observations or the uptake of habitat monitoring into farm management protocols 13 (Section 3.7).

| Surveying appropriate reference sites and environmental variables
Ideally, to fully assess the environmental effects of an aquaculture system, surveys should be conducted before farms are established in any given area, and then compared to results seen during and after implementation, as well as at control sites, thereby adopting a before-after-control-impact (BACI) design. 262Beyond BACI studies may often include multiple, additional control sites away from the farm, which experience similar background environmental conditions to the farm site (e.g.depth, sediment, hydrology) but are at an appropriate distance away so as to not be affected by the presence of aquaculture species (e. should also be compared to areas with similar environmental conditions where there is no structural habitat, as this is where farms are often implemented.Factoring in the monitoring of other variables such as macroalgal species and biomass, depth, light, temperature, sediment and water biogeochemistry (e.g.]105 To assess the effect of habitat provisioning beyond the direct footprint of a cultivation site, reference sites can be set at incremental distances away from the site to determine a sphere of influence or radius of attraction for species (e.g.Ref. 124).In many aquaculture sites, wider effects can be relatively small, e.g.limited to <50 m for epibenthic macrofauna in Canadian blue mussel farms. 124For different taxonomic groups however, it is likely that the influence of cultivation sites will extend further outside of the farm's footprint, such as for larger megafauna that may be deterred from foraging grounds, as observed with dolphins and shellfish aquaculture in New Zealand.

| Species metrics to be taken and statistical analyses of habitat value
Quantifying the habitat value of macroalgal cultivation sites does not only rely on determining the species present in the site and their abundances but also on understanding their usage of and behavioural interactions with the site at various life stages in the long term, as well as monitoring their physiological condition and fitness (e.g.juvenile recruitment).Juvenile recruitment success underpins population fitness and biodiversity, so it should be monitored within cultivation sites and surrounding areas to address whether macroalgal farms enhance wild populations through juvenile recruitment or simply aggregate individuals from surrounding areas.Quantifying the size and nominal age of individual organisms is also needed to understand juvenile recruitment. 8,124Biomass measurements are also important to determine nutrient and energy flows in ecosystem-wide and food web models 120 (see Section 3.5), but can be estimated for wild fish using published length-weight conversions (e.g.ecoCEN 265 ).Many of the methods outlined in methods are also often necessary to determine the behavioural activity of species at macroalgal cultivation sites, for example, feeding, breeding and sheltering or avoidance of the site.Tagging or biologging species also allows the behaviour of individuals to be monitored and site fidelity to be established. 238Feeding behaviour can also be determined through gut content or isotopic analysis (e.g.Ref. 266), which can establish food web effects of the farm.Better knowledge of species' activity and feeding behaviours will help determine what attracts them to farming infrastructures, and the significance of the farm habitat.This will furthermore help to improve the understanding of how farming practices can be better designed to maximise the habitat value of macroalgal cultivation sites and mitigate disturbance on key life stages of inhabiting species, through EAA.
The use of diverse biotic indices and statistical approaches can also provide various insights into overall community and ecosystem health, determining the habitability of an area for different species.
For example, to quantify infaunal biodiversity, a large variety of statistical approaches 183,267,268 and biotic indices are used (summarised in Refs.117,269).Indices lend themselves to standardisation and they can be fine-tuned to detect certain types of pressure, for example, the Infaunal Trophic Index (ITI), which takes into account species sensitivity to organic enrichment 270 could be used to quantify the impact of organic exudates and cast-off from macroalgal farms.The Infaunal Quality Index (IQI) is also used to assess sediment quality and would be useful for detecting disturbance below macroalgal cultivation sites. 2692][273] The choice of data analysis tools is therefore important and should be considered carefully when designing habitat value surveys to ensure they fulfil assessment objectives and are comparable to other sites.
For comparability of habitat value between cultivation sites and aquaculture species, a standardised set of variables, biotic indices and statistical approaches should be produced, which would enable better quantification of the habitat value of macroalgal cultivation. 1175][276] MarESA can be applied to monitoring different forms of aquaculture and their impacts in terms of magnitude, extent, duration and frequency of the effect, so that pressures from different activities can be compared on an equal footing.

| Modelling the wider ecosystem effects of habitat provisioning for an ecosystem approach to aquaculture (EAA)
Despite the increasing number of survey methods available for quantifying habitat provisioning in macroalgal cultivation sites, the resources available to conduct such field surveys are limited, particularly at small-scale farms and they tend to focus within a farm's footprint.
Models may be used instead to predict ecosystem-wide effects based on established relationships with primary productivity, nutrient and energy flows and readily available environmental and species data.
At macroalgal cultivation sites, models have already been run for determining effects on plankton assemblages at different farm scales (e.g.Refs.65,67,71,73,74) and ecosystem-wide effects on food web dynamics (e.g.Ref. 120).Models of carrying capacity in aquaculture systems assess the maximum production potential of a cultivated species that can be supported in an area based on environmental conditions, optimal stocking density, cultivation approaches and environmental impact. 72,277Carrying capacity models can be used in the development of EAA to assess ecosystem impacts beyond the direct footprint of the cultivation site and ensure sustainability. 277,278In Sanggou Bay, China, carrying capacity models have been used to assess production limits of cultivated kelp and oysters based on species growth dynamics and environmental data 279 ; however, these did not focus on the ecological carrying capacity of the site, rather on maximising production.Ecological carrying capacity in macroalgal cultivation sites should be investigated further to balance production with ecosystem management goals, as has been outlined for shellfish and finfish cultivation previously. 73,277,278though models rely on initial species input data from field surveys to be calibrated, ground truthed and verified, data from other prepublished studies and reports can also be used. 120As ecosystems are complex, models can tend to oversimplify aquacultureenvironmental relationships 71 ; however, as our understanding of these systems improves, the usefulness of habitat provisioning models will also increase.Models can also be used to assess cumulative ecosystem impacts of IMTA systems or multipurpose platforms such as integrated aquaculture and marine renewable energy sites to inform marine spatial planning and policy related to EAA. 280

| Data distribution and access
In order to inform decisions on optimising the habitat value of macroalgal cultivation sites in EAA, high quality comparable data from multiple sites will need to be used.To facilitate this, global standardised data sets should be generated with available data on habitat provisioning in macroalgal farms or other aquaculture types.
Conceptual frameworks such as the Essential Biodiversity Variables (EBVs) 281 could aid in creating interoperable data sets based on data collected using common methodologies.These frameworks could then be made available in open-access repositories to facilitate habitat value or biodiversity assessments. 282

| Integration of monitoring techniques into farm management protocols and policy
Currently, policy relating to macroalgal cultivation at either national or international level is not well established, and farm management protocols are often focused on reducing waste, pollution, disease outbreaks and damaging epibionts, rather than maximising the ecological value of the site.Standardised quantification of habitat provisioning of macroalgal cultivation sites would be greatly facilitated through the creation of clear monitoring criteria and guidance from regulatory and accreditation bodies through farm management protocols or policy. 5,13,107Monitoring techniques should adhere to licencing and certification standards, whilst also benefitting farmers to incentivise their usage, for example, via optimising harvest for non-compliance. 72The environmental data currently reported relates mostly to water quality and pollution levels 72 ; however, reporting could be expanded to include biodiversity or habitat monitoring. 13To engage other stakeholders and facilitate social licencing, monitoring techniques could also be targeted at verifying potential economic benefits to other marine industries, including fisheries, g. Refs.100,101).Where macroalgal cultivation sites have already been implemented before baseline conditions were established, habitat value has often been compared to other reference habitats, such as seagrass beds (e.g.Ref. 129) or wild macroalgae populations (e.g.Ref. 105).To understand any added habitat value created by macroalgal cultivation sites to an area, farms schedules and increasing product yield and grade or facilitating regulation and licencing of farms.Several sustainability and organic certification standards for macroalgal cultivation production (outlined in Ref. 9) discuss the need for farmers to assess the positive and negative environmental impacts of their farms and establish sustainable management plans to enable their products to be accredited; however, very little direct guidance on evidence-based monitoring is given.For example, the Aquaculture Stewardship Council (ASC)-Marine Stewardship Council (MSC) sustainable seaweed standard for both wild and farmed seaweeds sets a number of requirements for farmers to demonstrate that they are actively reducing any potential negative environmental impacts of their farms, including on native species and habitats.283Elements of the ASC-MSC standard include habitat, ecosystem structure and function, species status, species management, waste management and pollution control, energy efficiency, disease and pest management practices and introduced species management.The ASC-MSC standard does not, however, detail any specific monitoring techniques required to provide evidence for this.If certification standards could encourage this form of standardised data collection as a requirement for certification, it would greatly incentivise farmers to integrate habitat provisioning monitoring techniques into their current farm management protocols.Standardising methods for quantifying habitat provisioning and the associated ecological (and economic) value is undoubtedly more challenging than for data collection related to assessing and managing impacts around waste management and pollution control, energy efficiency, disease and pest management practices.However, encouraging monitoring of habitat provisioning is essential for realising an EAA and assuring the sustainability of the industry.72Marine licencing bodies could also set environmental and habitat monitoring as a legal requirement to grant farming permissions, particularly in newly emerging regions of macroalgal cultivation, such as Europe.The EU strategic guidelines for a more sustainable and competitive EU aquaculture in 2021-2030 highlights the limited data reported on environmental indicators related to aquaculture and the need to obligate farmers to monitor and report environmental data to licencing systems and for regulating bodies to enforce sanctions

TA B L E 1
List of suitable monitoring techniques to assess habitat provisioning in macroalgal cultivation sites for different taxonomic and functional groups.Potential pros and cons for each technique are discussed with recommendations for improved use and examples of previous usage in macroalgal cultivation sites and other similar habitats

it can be used for Pros (+) and cons (−) for use in macroalgal cultivation sites to assess habitat provisioning Recommendations for future use Example uses in macroalgae
but portable sensors and combined modelling methods may aid in development (−/+) Commercial monitoring systems are expensive for small-scale farmers, but low-cost options are being developed (e.g.Refs.170,171) (−/+) Requires good connectivity for instantaneous data transfer i.e. 4G networks, which may not be available in all farm locations (−) Should standardise sensors used and their placement on farms Should invest further research and development into these systems to streamline data connectivity and lower production and running costs for use in small-scale farms Should ensure systems are simple to use for operation and maintenance by farmers Microorganisms: e.g.Refs.172 (O); 169 (review) Plankton: e.g.Refs.170,171 (S) (Continues) Monitoring technique & species group(s) (M), finfish (F), shellfish (S) aquaculture, wild macroalgal populations (W) or other (O) Collecting species on macroalgal biomass Microorganisms and epibionts Easy to collect samples on farm maintenance or harvest trips (+) Species can be then identified by several methods e.g.: • Microscopy or by eye -time-consuming and requires taxonomic expertise, but captures total biodiversity at a given time point above a visible size class (−/+) • Molecular methods (see below) -faster but requires established reference libraries (+/−) • Photographic analysis (e.g. using ImageJ or equivalent software) -standardises measurements for sessile epibionts, but excludes less abundant mobile species, and those obscured in holdfasts (+/−) Should standardise methods for collection, preservation, species identification and enumeration Should assess all parts of the plant and account for habitat volume of holdfasts or surface area of blades and stipes, as species richness scales with plant size. 173These methods are described in Refs.88,105 Should combine image analysis and taxonomic identification (e.g.Ref. 83) for increased accuracy Microorganisms: e.g.Refs 55,174-177 (W) Epibionts: e.g.Refs 88,101-105 (M);83,173,178,179 (W) Constrained by depth and substrate type, e.g.sampling offshore farms or those suspended over bedrock is challenging (−) Species can be then identified via: • Taxonomy -expensive, time-consuming and requires expertise (−) • Molecular methods (see below) -faster but requires established reference libraries (+/−) Should conduct pre-surveys via side-scan sonar or drop-down cameras to determine substrate types and any vulnerable or protected habitats to be avoided 5 Should use standardised monitoring methods, e.g.Refs.181,182, and diverse biotic indices and statistical approaches (e.g.Refs.117,183) Benthic microorganisms: e.g.Refs.184-186 (F) Benthic infauna: e.g.Refs.100,101,118,120,121 (M); 184 (F) Small beam trawls Benthic epifauna Routinely conducted and replicable around different benthic sites and effective at assessing total community assemblages (+) Relatively destructive and may be hard to conduct around farm infrastructure or directly below farms (−) High levels of settled farm detritus may clog nets and hinder accurate quantification of epifauna (−) Can subsequently determine biometrics, age/size and gut contents of epifauna (+) Should standardise methods (e.g.tow size, net size, tow length and duration, sample preservation, species identification and enumeration).Where trawl sizes differ, data should be standardised by area/volume Static sampling methods (e.g.trap or camera surveys) are more appropriate around aquaculture infrastructure Benthic epifauna: e.g.Refs.187 (S); 188 (F); 180 (O) Traps or nets Benthic epifauna and finfish Relatively simple to deploy below or around cultivation sites for crustaceans and fish species (e.g.Ref. 101) (+) Traps may be particularly effective for surveying species of commercial importance, but will be biased for certain species (+/−) Can subsequently determine biometrics, age/size and gut contents of caught individuals or tag them (see mark recapture or biologging below) (+) Should be used with other survey methods to derive whole-site biodiversity Should standardise methods (e.g.net or trap size, length of deployment, baited or unbaited).Where length of deployment differs among studies, data should be standardised to catch per hour Benthic epifauna: e.g.Refs.101 (M); 187 (S) Finfish: e.g.Refs.129 (M); 187,189 (S); 130 (S, F & O) TA B L E 1 (Continued) Monitoring technique & species group(s)

it can be used for Pros (+) and cons (−) for use in macroalgal cultivation sites to assess habitat provisioning Recommendations for future use Example uses in macroalgae
(M), finfish (F), shellfish (S) aquaculture, wild macroalgal populations (W) or other (O) and may only capture species a set time point or time of day e.g.does not account for diurnal movements of organisms (−) Not suitable for benthic surveys in offshore farms that may be beyond diving depth limits (−) Divers presence may cause additional disturbance to mobile species affecting counts (−) Should standardise methods for surveys e.g.either point counts, transect lengths and widths or manta tows etc. and describe search effort and time of day Should use cameras where possible to increase replicability and reduce individual observer bias (see remotely operated camera surveys below) Monitoring technique & species group(s)

it can be used for Pros (+) and cons (−) for use in macroalgal cultivation sites to assess habitat provisioning Recommendations for future use
TA B L E 1 (Continued)Monitoring technique & species group(s)

it can be used for Pros (+) and cons (−) for use in macroalgal cultivation sites to assess habitat provisioning Recommendations for future use
Table 1 are capable of determining size and age classes of individuals, through the direct measurements of captured, from photographed organisms using a scale, or through visual estimates from experienced divers or observers (e.g.Ref. 200).Visual